The overall objective of experiments outlined in this proposal is to develop methods for generating EBV mutant viruses. Only through construction and analysis of such mutants will it eventually become clear which EBV genes are essential for the process of lymphoid cell immortalization, epithelial cell transformation, and oncogenesis. The experiments rely on three different approaches. In the first approach naturally occurring evolutionary variants will be characterized which affect the entry pathway from latency into viral replication in lymphocytes. Several new variants which affect genome size, configuration, and the structure of orilyt are in hand. These will be studied in detail. The second approach will attempt to take advantage of newly created cell lines which have a virus receptor and are also permissive for virus replication. These cells will be the substrate for recombination between EBV strains or between subgenomic DNA fragments of one strain and another complete genome. The third approach, for which we have the most preliminary data, explores the use of recombinogenic cassettes in which a selectable marker (for drug resistance) is flanked by EBV sequences, The objective is to perfect a general method of EBV mutagenesis by homologous recombination in which EBV genes are inactivated by insertion of a selectable marker. The proposed experiments are based on combined biologic and genetic approaches which should eventually increase our understanding of how EBV genes and gene products interact to produce such diverse pathologic outcomes.